Binding in Radical-Solvent Binary Complexes: Benchmark Energies and Performance of Approximate Methods.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Chemical Physics

Background:

• Weak interactions between radicals and their environment significantly impact their properties and reactivity.
• Understanding these interactions is crucial for predicting chemical behavior in various systems.

Purpose of the Study:

• To compute benchmark binding energies for radical-solvent complexes using high-level ab initio methods.
• To evaluate the performance of modern Density Functional Theory (DFT) functionals for these interactions.
• To identify systematic errors in DFT functionals when describing radical-solvent interactions.

Main Methods:

• High-level ab initio coupled cluster theory (up to CCSDT(Q)) with core correlation and relativistic corrections was employed.
• Benchmark binding energies were computed for 12 binary complexes involving neutral and charged small radicals.
• Radicals were complexed with polar solvent molecules: water (H2O) and hydrogen fluoride (HF).
• The performance of various DFT functionals was assessed against these accurate ab initio results.

Main Results:

• Radical hydrogen-bonded complexes were generally well-described by most DFT methods.
• Two-center-three-electron interactions were found to be systematically overbound by most evaluated DFT functionals, including range-separated ones.
• Electron-rich metal-water complexes did not exhibit such systematic errors.
• No single DFT functional achieved chemical accuracy across all tested radical-solvent complex types.

Conclusions:

• Most DFT functionals show limitations in accurately describing radical-solvent interactions, especially two-center-three-electron bonds.
• Specific types of radical-solvent complexes, like electron-rich metal-water interactions, are better predicted by current DFT methods.
• Further development of DFT functionals is needed to achieve reliable accuracy for diverse radical-solvent interactions.